The purpose of this study was to determine the extent of transmission of tuberculosis in a large prison population over an 18-mo period. Restriction-fragment-length polymorphism (RFLP) analysis of isolates of Mycobacterium tuberculosis was performed, using the insertion sequence IS6110 and the plasmid pTBN12. Patients infected with strains having the same fingerprint were grouped in clusters. Medical records were reviewed and movement of inmates among prisons was examined for selected patients. Tuberculosis was diagnosed in 216 inmates (case rate = 2,283 per 100,000 per year). Isolates from 210 (97%) patients were fingerprinted, 155 (74%) were grouped in 25 clusters, and 55 (26%) showed a unique fingerprint. Recent infection was inferred in 62% of these patients. Eighty-four percent (161 of 192) of patients tested were human immunodeficiency virus (HIV)-positive, of whom 121 were in clusters and 40 were not (p = 0.74). Patients in clusters were less adherent with tuberculosis treatment than those not in clusters (p < 0.05), and prison transmission of resistant strains was observed. It is crucial that infection control guidelines be fully implemented in the prison setting to prevent tuberculosis transmission.

This study provides evidence that HIV-infected inmates living under conditions of high environmental infectivity can be infected with two different strains of M. tuberculosis. This finding has implications for the tuberculosis-control programs in prison.

Genetic alterations in the rpoB gene were characterized in 50 rifampin-resistant (Rif r ) clinical isolates of Mycobacterium tuberculosis complex from Spain. A rapid PCR-enzyme-linked immunosorbent assay (ELISA) technique for the identification of rpoB mutations was evaluated with isolates of the M. tuberculosis complex and clinical specimens from tuberculosis patients that were positive for acid-fast bacilli (AFB). Sequence analysis demonstrated 11 different rpoB mutations among the Rif r isolates in the study. The most frequent mutations were those associated with codon 531 (24 of 50; 48%) and codon 526 (11 of 50; 22%). Although the PCR-ELISA does not permit characterization of the specific Rif r allele within each strain, 10 of the 11 Rif r genotypes were correctly identified by this method. We used the PCR-ELISA to predict the rifampin susceptibility of M. tuberculosis complex organisms from 30 AFB-positive sputum specimens. For 28 samples, of which 9 contained Rif r organisms and 19 contained susceptible strains, results were concordant with those based on culturebased drug susceptibility testing and sequencing. Results from the remaining two samples could not be interpreted because of low bacillary load (microscopy score of 1؉ for 1 to 9 microorganisms/100 fields). Our results suggest that the PCR-ELISA is an easy technique to implement and could be used as a rapid procedure for detecting rifampin resistance to complement conventional culture-based methods.Tuberculosis is a major health problem not only in developing countries but also in the developed world among high-risk groups, such as human immunodeficiency virus-infected patients, immigrants from countries in which tuberculosis is endemic, the homeless, and prison inmates (3,20). The recent resurgence of tuberculosis in developed countries has been accompanied by an increase in the prevalence of drug-resistant disease. Resistance to rifampin is of particular importance since it is a marker for strains of multidrug-resistant tuberculosis that have been associated with outbreaks of disease in hospitals, prisons, and other institutions worldwide (2, 3, 5). The rate of rifampin resistance among strains of Mycobacterum tuberculosis isolated in Spain varies from 29% in patients who have previously received treatment (1) to 1.6% in those who have not received any prior antimycobacterial therapy (9).Ninety-six percent of rifampin-resistant (Rif r ) strains of M. tuberculosis possess genetic alterations within an 81-bp fragment of the rpoB gene that encodes the ␤ subunit of DNAdependent RNA polymerase (18, 26). Many different allelic variations have been detected within this region (11,13,18,21), and specific rpoB genotypes are known to be associated with high-level rifampin resistance (15, 26). Molecular assays that have been used to screen the rpoB gene for rifampin resistance mutations include DNA sequencing (11), heteroduplex analysis (27), PCR single-stranded conformational polymorphism (21, 22), line probe assay (4, 6), mismatch analysis (14), and real-...

In order to determine the clinical significance of mixed oropharyngeal candidiasis (Candida albicans plus a non-albicans strain of Candida) in patients infected with HIV-1, a retrospective chart review was done in 12 HIV-1-infected patients with a clinical episode of oropharyngeal candidiasis, in whom a mixed culture of Candida albicans (found to be fluconazole-sensitive) plus a non-albicans species of Candida was obtained from their oral cavities. This group was compared with 26 HIV-positive patients (control group) with oropharyngeal candidiasis due to Candida albicans (found to be fluconazole-sensitive). Antifungal susceptibility testing was performed by a broth microdilution test with RPMI-2% glucose. A fungal strain was considered fluconazole-sensitive if its MIC was < 0.5 micrograms/ml. Both the study and control groups had similar clinical and demographic characteristics. All the patients were severely immunocompromised, with a mean CD4+ lymphocyte count of 63/mm3 (95% CI 41-84) and 80/mm3 (95% CI 25-135) in the study and control groups, respectively. In the study group, seven patients had Candida albicans and Candida krusei in their oral cavity, four had Candida albicans and Candida glabrata, and one had Candida albicans and Candida tropicalis. Antifungal therapy consisted of ketoconazole (5 patients in the study group, 14 in the control group) or fluconazole (7 patients in the study group, 12 in the control group); no statistically significant difference in clinical outcome was observed. Fungal strain persistence after therapy was frequently observed in both groups. It is concluded that non-albicans strains of Candida, less sensitive to azole drugs than their Candida albicans counterparts, are not clinically relevant in episodes of mixed oropharyngeal candidiasis in HIV-1-infected patients.

To assess the clinical significance of splenic tuberculosis in patients infected with human immunodeficiency virus (HIV) type 1, we compared 20 patients who had splenic tuberculosis with 20 randomly selected, HIV-infected patients with culture-proven tuberculosis for whom splenic involvement had been ruled out by ultrasonography. All of the patients were male prison inmates and intravenous drug users. Statistically significant differences (P < .05) were detected between patients with splenic involvement (median CD4+ cell count, 54/mm3) and those without splenic involvement (median CD4+ cell count, 92/mm3). No specific symptoms suggesting splenic involvement were detected in the patients with splenic tuberculosis. All patients received antituberculous drugs, and none of these patients required splenectomy. The median survival was similar in both groups. Splenic tuberculosis occurs in more-severely immunocompromised HIV-infected patients, the prognosis is generally good, the clinical response to therapy is usually favorable, and splenectomy is rarely necessary.